Research And Implementation Of Wireless Wideband Signal Compression Sensing Technology

Compressive Sensing(CS)is one of the most popular topics in signal processing field,which is used to process signals that are sparse in some specific basis,it provides advantages of sampling original signal in low rate and economizing storing memory.Thus CS makes an important influence on scenes that sampling rate and storing memory are limited.This paper is mainly concerned about the theory and hardware implementation of Wide-band sparse signal using CS.We do research on adaptive signal reconstruction based on a classical compressed structure called Analog-to-Information Conversion(AIC)[1].Besides,this paper summarizes and compares two kinds of main compressed sensing sampling structure models:(AIC)and Modulated Wideband Converter(MWC)[2].Both models involve mixing original signal with Nyquist sampling speed pseudo random signal which turns out to be anther bottleneck in CS.To overcome the challenge of high speed mixing,this article studies on non?mixing CS which abandons the mixing part.The main contributions of this article are as follows:1.There exists transition relationship between adjacent time windows in AIC model and reconstructing wideband sparse signal with Kalman filter iteration has been proved to be a good method.However,the iteration process requires a prior knowledge of noise covariance matrix and the sparsity of signal which are unknown or time-varying in actual scenery.Providing that both noise covariance matrix and sparsity are unknown,this article puts forward a corresponding adaptive sensing algorithm and the simulation results show that the proposed adaptive reconstruction algorithm can almost get the same good performance as algorithms with prior knowledge of the noise covariance and signal sparsity.2.This paper designs a complete remote data acquisition system based on high-speed ADC-AD9361[3]and wideband hardware platform ZC706[4],software component for this system is also provided.3.A compression system based on MWC model is designed in this paper after investigating on hardware implementation of CS.Considering the difficulties of high speed mixing,this article proposes a budget of CS system construction program from the perspective of economizing based no mixing CS,which gives guidance for building an economy CS system.